Use of marijuana (cannabis) among pregnant women in the US is increasing with prevalence as high as 14% among 12?18 year old pregnant women. The American College of Obstetrics and Gynecology recommends that pregnant women avoid marijuana due to evidence that it affects the fetus and may interfere with brain development. Studies in animals appear to support this recommendation. Although other constituents of marijuana cannot be discounted, the general scientific consensus is that ?9-tetrahydrocannabinol (THC), the most abundant and psychoactive component in marijuana, is the likely perpetrator of the developmental neurotoxicity of marijuana. THC can dysregulate cannabinoid receptor 1 signaling during pregnancy and can result in adverse outcomes such as impaired fetal brain development, lower birth weight, increased fetal resorption, and even in utero deaths. THC can impact axon growth in the developing mouse fetal brain. Chronic exposure to THC leads to long-term behavioral deficits in male adolescent mice, akin to those observed in schizophrenia. However, these rodent and in vitro studies were conducted at high THC doses or concentrations and therefore their applicability to humans, where THC plasma concentrations are sub- micromolar, is unknown. For many reasons, observational clinical studies in pregnant women who use marijuana are not informative as to whether marijuana is safe when used during pregnancy. Due to the limitations of all the above approaches, we propose here a systems pharmacology approach to begin to address this significant public health question. Through data obtained by this project and Projects 1 & 2, we will predict and then verify the magnitude of maternal-placental-fetal exposure to THC and its psychoactive metabolite, 11-OH-THC, throughout pregnancy, after both oral and inhalational (smoking) use of marijuana. To do so, we will refine and extend a novel maternal-fetal Physiologically Based PharmacoKinetic (m-f-PBPK) model we have developed. In addition, in an exploratory manner, we will determine whether these cannabinoids produce any molecular signatures indicative of short or long-term developmental neurotoxicity in humans. Our approach uses novel and innovative tools (e.g. m-f-PBPK model, development of an inhalational m-f-PBPK model, quantitative targeted proteomics, transcriptomics, proteomics and metabolomics) to address a compelling public health question. 1